智能手机App与数字半球摄影测量叶面积指数对比研究

基于自主开发的智能手机App（LAISmart）对针阔混交林、阔叶林和农作物3种植被类型的叶面积指数（leaf area index, LAI）进行测量，并以数字半球摄影（digital hemispherical photography，DHP）的测量结果作为参考值进行对比分析．结果表明，虽然LAISmart与DHP的LAI值总体上具有高度一致性（R2=0.95，RMSE=0.68），但是，LAISmart的性能受到植被叶片密集程度的影响．研究发现:LAI>3.9时，LAISmart的测量结果会明显低于DHP的测量结果；智能手机成像传感器的自动曝光模式，是引起LAISmart在测量LAI高值区域估值偏低的重要影响因素；当对LAI高值区域的LAISmart图像进行降低曝光度处理后，LAISmart和DHP的测量结果偏差得到进一步降低，且LAISmart测量结果的精度可以提高49%左右．此外，LAISmart的较窄视场角几乎不会对其测量结果产生影响，若能在调节智能手机曝光度的条件下使用LAISmart，则具有更高效率和更低成本优势的LAISmart可以成为替代DHP的有效方法． 

Comparison of Smartphone App and digital hemispherical photography for Estimating Leaf Area Index

Leaf area index (LAI) is a structural parameter that describes distribution characteristics of vegetation leaves in canopy．LAI can affect redistribution of solar radiation, photosynthesis capacity of vegetation and microclimate in vegetation growth environment. Therefore, obtaining LAI in ground measurements is of great significance to understand growth status of vegetation．Digital hemispherical photography (DHP) has been widely adopted due to its simplicity and low cost．In recent years, smartphone-based LAI measurement methods have been extensively used．However, in-depth comparison and verification between smartphone and DHP methods have not been done．Performance of smartphone for measuring LAI needs to be validated and its potential to replace DHP needs to be explored．In this paper we measured coniferous-broadleaved mixed forest, broadleaved forest and crop by self-developed smartphone APP (LAISmart)．These DHP measurements were used as reference for comparative studies．LAI values derived from LAISmart were found highly consistent with DHP (R2=0.95, RMSE=0.68), but performance of LAISmart was affected by density of vegetation leaves．When LAI was > 3.9 , LAISmart significantly underestimated measurements of DHP．Auto-exposure mode of smartphone imaging sensor was an important factor for LAISmart underestimations in LAI in high-value areas．When exposure of LAISmart images in LAI high-value areas are reduced, deviation between LAISmart and DHP was reduced, accuracy of LAISmart was improved by 49%．Narrow sensor field of view (FOV) of LAISmart has little effect on measurement results．These data suggest that LAISmart, with higher efficiency and lower costs, can be an alternative to DHP．