基于ImageJ机器学习技术的文物建筑表层劣化统计分析及应用研究

Statistical Analysis and Application Research on Surface Degradation of Historical buildings Based on ImageJ Machine Learning Technology

受自然界风雨侵蚀的长期作用,文物建筑本体的表层劣化往往是其最主要的病害类型之一,在此类文物保护修缮方案的制定过程中,经常需要对其表层劣化的程度进行量化统计分析,作为综合判断其成因的重要参考指标,为合理制定保护修缮措施提供依据。通常情况下,文物建筑本体的表层劣化迹象大多呈极不规则状分布,仅靠人工测量的方式很难准确获取各劣化残损点的面积、周长及空间分布状态等参数,难以为后续各项工作的有效开展提供准确的基础数据支撑。基于ImageJ的机器学习技术能够通过一定的图像训练技术实现对特定范围内图像基于光谱感知差异的有效识别,并可对识别结果进行面积、周长计算等诸多统计分析,将此项技术应用于砖石类文物建筑表层劣化迹象的准确统计分析过程,能够得到令人满意的结果。本文以文物建筑外墙表层残损状况量化分析为例,初步探索此项技术的使用路径,并对分析结果及应用场景进行评判和展望。

For Historical buildings,the surface deterioration of cultural relics is often one of the most important disease types due to the long-term effect of wind and rain erosion in nature.In the process of making protection and repair plans for such cultural relics,it is often necessary to make quantitative statistical analysis on the degree of surface deterioration as an important reference index to comprehensively judge its causes and provide a basis for making reasonable protection and repair measures.Usually,the surface deterioration signs of Historical buildings are mostly irregularly distributed,so it is difficult to accurately obtain the parameters such as the area,perimeter and spatial distribution of each deteriorated damaged point only by manual measurement,and it is difficult to provide accurate basic data support for the effective development of subsequent work.The machine learning technology based on ImageJ can effectively identify the images in a specific range based on the difference of spectral perception through certain image training technology,and can carry out many statistical analysis on the identification results,such as area and perimeter calculation.When this technology is applied to the accurate statistical analysis process of the signs of surface deterioration of the brick and stone heritage buildings,satisfactory results can be obtained.In this paper,the quantitative analysis of the damage of the external wall surface of a cultural relic building is taken as an example to explore the application path of this technology,and the analysis results and application scenarios are evaluated and prospected.