基于机器学习的天文光电图像特征细节识别研究

Feature detail recognition of astronomical photoelectric image based on machine learning

为了提高天文光电图像特征检测识别能力,需要进行天文光电图像特征细节识别,提出基于机器学习的天文光电图像特征细节识别方法。建立天文光电图像特征细节识别模型,在大气散射环境下进行天文光电图像的热敏感强度自适应融合,采用模板匹配技术进行天文光电图像的信息增强处理,采用大气散射特征点匹配方法进行天文光电图像的细化滤波处理,采用大气散射特征点特征检测方法进行图像特征提取,使用亮度分量进行天文光电图像特征细节透射分析,对提取的图像细节特征量采用多尺度机器学习方法进行天文光电图像红外探测和预测,实现天文光电图像特征细节识别。仿真结果表明,采用该方法进行天文光电图像特征细节识别的精度较高,图像细节特征分辨力和准确性较好。

In order to improve the recognition ability of astronomical photoelectric image feature detection,it is necessary to carry out the feature detail feature recognition of astronomical photoelectric image,and a method of feature detail recognition is proposed based on machine learning.An image feature detail recognition model is established,the thermosensitive intensity of the astronomical image is adaptively fused in the atmospheric scattering environment,the information enhancement of the astronomical photoelectric image is processed by the template matching technique,the fine filtering of the astronomical photoelectric image is carried out by the atmospheric scattering feature point matching method,the image features are extracted by the atmospheric scattering feature point feature detection method,the brightness component is used for the transmission analysis of the astronomical photoelectric image feature details,and the multi-scale depth learning method is used to detect and predict the image details of the astronomical photoelectric image.The simulation results show that the method has high precision and good resolution and accuracy of image features.