摘要
应用图像分析技术研究了不同水分条件下棉花冠层光谱对水分变化的响应。比较了拍摄角度(包含顺光和逆光)、阴天和晴天、拍摄时间、品种及数码图像的像素模式等因素对图像质量的影响,以期建立标准的近地面数码图像获取方法。研究结果显示,品种、天气状况及像素模式对光谱参数的影响远小于拍摄角度、顺逆光条件和时间对其的影响,选择在中午阳光直射时拍摄图像可以减小光照条件的影响;在不同水分条件下分析发现,低水处理棉花叶片对蓝光的吸收增强,但对红光的吸收减弱,且对绿光的反射变弱。因此,图像分析技术可以作为棉花干旱诊断的辅助指标。
Image analysis were used for analysing cotton canopy spectrum response to soil humidity changes. Imagetaking angle, radiation density, timing and crop varieties and images format were taken into account in searching a optimum image-taking conditions and to build up a standard image-taking method. Our study indicated that crop varieties and radiation have much less effect on image quality as compared to image-taking angle and timing, and we suggest that 11：00 - 13：00 be the best time for image-taking. The study also found that cotton canopy tends to absorb more blue color radiation but absorb less red and green color light. This results indicated tthat image analysis could be used as an assistant index for cotton crop drought diagnosis.
出处
《石河子大学学报（自然科学版）》
CAS
2009年第5期541-547,共7页
Journal of Shihezi University(Natural Science)
基金
国家自然科学基金(30560069)
国家科技支撑计划(2007BAC20B03)
农业部公益性行业科技专项(200803030)
