Yapeng Wu

Radiometric Calibration (RC) is a critical process in aerospace infrared remote sensing that establishes the relationship between the radiation energy of observed objects and the Digital Number (DN) output from sensors, which is fundamental for ensuring high-precision applications of infrared remote sensing data. At present, Source-Based RC (SBRC) is the predominant method, relying on a variety of Radiometric Sources (RS) including in-orbit blackbodies, or natural targets such as lakes, oceans. This approach, while effective, imposes constraints on remote sensing systems such as space & weight allocation for RS and additional observation time for RC. Moreover, the reliance on physical calibration sources can introduce uncertainties due to factors such as imperfect emissivity of in-orbit blackbodies, lack of data consistency due to varied RS types, and variations in environmental conditions. In this paper, we propose a novel RC method named Artificial Intelligence Radiometric Calibration (AIRC), which directly generates RC coefficients for the in-orbit remote sensing satellites using the physical and environmental parameters of the sensor. We first theoretically prove that RC coefficients can be derived as functions of the sensor states. Next, we propose our Neural Networks for infrared Radiometric Calibration (RCNN), to learn this relationship based on historical high-accuracy calibration data, enabling a shift from Reference Traceability (RT) to States Traceability (ST). Then, to verify the feasibility of the proposed scheme, we train and test an Multi-layered Perceptron (MLP) as a simple implementation of RCNN based on our long-term well-curated RC data from our FengYun-4A Avanced Geosynchronous Radiation Imager (FY-4A AGRI), and the experiments show that the proposed method achieves high-accuracy RC comparable with the official RC method applied on FY-4A AGRI that uses an in-orbit blackbody. Our study showcases how to conduct RC using the “reason (the states of sensor) - results (calibration coefficient)” logic, as supplement to the existing “result (observation to RS) - reason (calibration coefficient)” logic, which promotes constituting a uniform observing system for cross-platform infrared satellites.

Radiometric Calibration (RC) is a key step in high-precision aerospace infrared remote sensing. Currently, source-based RC is the most prevailing methodologies, where a radiometric source such as blackbody with known radiation energy is observed by the sensor and the corresponding output digital number is recorded. However, the methodology is limited by several constraints, e.g., space and weight requirement for in-orbit blackbodies, and lack of uniformity and ideal properties of blackbodies. This paper presents a new artificial intelligence based RC that directly generates RC coefficients given the state of the sensor parameters. First, we derive the relationship between the RC coefficients and the physical parameters and environmental parameters of the sensor. Next, we steadily collect a large quantity of remote sensing data using our in-orbit satellites since 2018 and correspondingly train a neural network to fit such function for the generation of RC coefficients. Finally, our extensive experiments show that the proposed method achieves high-accuracy RC comparable with state-of-the-art methods that use in-orbit blackbody. To the best of our knowledge, we are the first to demonstrate an artificial-intelligence-based RC method for satellites without observation to blackbodies during in-orbit remote sensing. We anticipate that the new approach could promote the accuracy and uniformity of global infrared remote sensing data, thereby promoting the ability to discern natural laws and understand longterm environmental or celestial trends. Besides, it can substantially save the cost for carrying and observing in-orbit blackbodies.

As the core component of permanent magnet motor, the magnetic tile defects seriously affect the quality of industrial motor. Automatic recognition of the surface defects of the magnetic tile is a difficult job since the patterns of the defects are complex and diverse. The existing defect recognition methods result in difficulty in practical application due to the complicated system structure and the low accuracy of the image segmentation and the target detection for the diversity of the defect patterns. A self-supervised learning (SSL) method, which benefits from its nonlinear feature extraction performance, is proposed in this study to improve the existing approaches. We proposed an efficient multihead self-attention method, which can automatically locate single or multiple defect areas of magnetic tile and extract features of the magnetic tile defects. We also designed an accurate full-connection classifier, which can accurately classify different defects of magnetic tile defects. A knowledge distillation process without labeling is proposed, which simplifies the self-supervised training process. The process of our method is as follows. A feature extraction model consists of standard vision transformer (ViT) backbone, which is trained by contrast learning without labeled dataset that is used to extract global and local features from the input magnetic tile images. Then, we use a full-connection neural network, which is trained by using labeled dataset to classify the known defect types. Finally, we combined the feature extraction model and defect classification model together to form a relatively simple integrated system. The public magnetic tile surface defect dataset, which holds 5 defect categories and 1 nondefect category, is used in the process of training, validating, and testing. We also use online data augmentation techs to increase training samples to make the model converge and achieve high classification accuracy. The experimental results show that the features extracted by the SSL method can get richer and more detailed features than the supervised learning model gets. The composite model reaches to a high testing accuracy of 98.3%, and gains relatively strong robustness and good generalization ability.