固液混合物体热成像图的识别处理方式研究

热成像图能够反映物体辐射的性能,可在图像处理后应用于固体与液体材料的识别,而固液混合物热成像图的图像处理是技术难点。文中旨在从固液混合物的热成像图中提取内部特征及轮廓特征,先进行包含线性灰度变换、直方图修正、PCA分析的不同顺序及组合方式的图像修正处理,后进行包含MSER特征点与Prewitt特征边界的提取,最终对处理后的图像进行比较分析,得到三条图像处理路径,分别可用于固体成分分析、液体成分分析和物体外部轮廓分析。该处理方式可用于垃圾分类与工业物料识别,具有广阔前景。

Recognition processing method of solid-liquid mixture thermal imaging

The infrared thermal images reflect the radiation performance of the object and can be applied to the identification of solid and liquid materials after image processing.The image processing of the infrared thermal images of the solid-liquid mixture is a technical difficulty.In order to extract internal features and contour features from the thermal image of the solid-liquid mixture,a combined image correction comparison experiment including linear gray scale transformation,histogram correction,and PCA analysis is performed firstly.And then,an extraction experiment including MSER feature points and Prewitt feature boundaries is performed.Finally,the processed images are compared and analyzed,and three image processing paths are obtained,which can be used for solid component analysis,liquid component analysis,and object external contour analysis.Treatment methods have broad prospects in material classification and identification.