Inference method for cause failure of sar equipment task based on double-layer nested structure

Abstract

The invention discloses a SAR equipment task failure cause reasoning method based on a double-layer nested structure, which relates to the technical field of SAR equipment support, including: collecting SAR image data of known abnormal types; reconstructing sample categories, extracting sample sets; calculating Image quality assessment features; image feature transformation, normalization processing, and class label merging; integration and acquisition of training data sets and training of two-layer models; processing unknown abnormal SAR images into test data sets; SAR equipment task failure reasoning. The present invention adopts a double-layer random forest model to combine the failure causes that are easy to be misclassified, thereby reducing the total number of categories, and uses image local features on the basis of the results of the first layer classifier to perform the second class on the samples that are easy to be misclassified. Secondary classification, thereby enhancing the accuracy of the random forest model for abnormal image classification, and at the same time overcoming the difficulty of using different terrain SAR image data to train the model is not effective, effectively improving the accuracy of SAR equipment task failure cause reasoning.

Description

technical field [0001] The invention relates to the technical field of SAR equipment support, in particular to a reasoning method for SAR equipment task failure causes based on a double-layer nested structure. Background technique [0002] At present, our army is in a period of transformation from a mechanized army to an informationized army, and the ability to collect, transmit and process information is increasing day by day. As a new type of efficient information acquisition weapon, SAR has become a new way of military observation and reconnaissance. SAR imaging is easily affected by many factors. When the SAR radar returns from a mission and cannot obtain a high-interpretation, good-effect, and clear enough image, that is, when the mission fails, it is necessary to perform fault detection and maintenance on the radar. . [0003] Traditional fault diagnosis and troubleshooting technologies mainly rely on expert experience, and have problems of poor reliability, low accu...

AI Technical Summary

Problems solved by technology

[0003] Traditional fault diagnosis and troubleshooting technologies mainly rely on expert experience, and have problems of poor reliability, low accuracy, and high false alarm rate, which affect the completion of daily drill tasks and even cause irreparable losses to actual military operations. These problems are serious Affects the logistic capability and combat performance of radar equipment, so there is an urgent need for a failure cause reasoning method oriented to mission completion

Images