期刊文献+

成像环境因素对烟叶图像采集结果的影响及校正研究 被引量:32

The Influences of Imagery Environmental Factors on the Acquisition of Flue-Cured Tobacco Leaf Images and Correction
下载PDF
导出
摘要 目的研究灯箱中采集烟叶图像时照明电压、不同品牌数码相机、聚焦距离及相机分辨率等成像环境因素对图像识别结果的影响,用相对颜色法对因成像环境因素影响而引起的烟叶图像颜色变异进行实时校正。方法通过试验筛选出颜色表现稳定的聚四氟乙烯材料作为颜色参照白板,在灯箱环境中,设置不同的因素处理,用数码相机同期采集烟叶与颜色参照白板图像,通过分析烟叶图像R、G、B颜色特征变化来确定成像环境因素构成的影响。提取烟叶图像三颜色分量与颜色参照白板对应颜色分量的比率,即相对颜色分量作为校正烟叶图像识别误差的特征参量。结果照明电压、不同品牌数码相机、相机分辨率相同而聚焦距离不同均使烟叶图像颜色产生变异,从而出现识别误差。结论相对颜色法能够实时校正识别误差,降低因成像环境因素影响而产生的颜色变异,为规范烟叶图像的数字采集环境奠定了基础。 [Objective] The study examined the influences of imagery environmental factors such as lighting voltages in a flue-cured tobacco leaf image capture lamp box, digital camera brands, focusing distances and resolutions on the result of image recognition. Relative color methods were employed to timely correct color variations caused by different imagery environmental factors. [Method] Based on the experiment, a white plane made of Teflon, for its color-stable characteristics, was selected to be the color reference plane. The images of both tobacco leaf under different treatments and its color reference plane were acquired in the lamp box. The influences of imagery environmental factors were clarified through the analysis of changing characteristics of R, G, B color vectors of tobacco leaf images. The relative color, defined as ratio of R, G, B color vectors against color reference planes was calculated. It was used as a parameter for the correction of color variations of tobacco images. [ Result] The results showed that lighting voltage in the lamp box, digital camera brand and given the same resolution, different focusing distances could cause color variations of tobacco leaf images, thus introducing the errors in recognition. [Conclusion] Relative color method was proved to be an effective approach to correct errors, lessen the influences of imagery environmental factors that caused the color variation, and lay a solid foundation for the standardization of shooting conditions for digital capture of tobacco images.
出处 《中国农业科学》 CAS CSCD 北大核心 2006年第12期2615-2620,共6页 Scientia Agricultura Sinica
基金 国家级星火计划项目(2002EA760003)
关键词 烟叶图像 成像环境因素 图像采集 颜色参照白板 Tobacco leaf image Imagery environmental factor Acquisition Color reference plane
  • 相关文献

参考文献21

  • 1Urena R,Rodríguez F,Berenguel M.A machine vision system for seeds germination quality evaluation using fuzzy logic.Computers and Electronics in Agriculture,2001,32(1):1-20.
  • 2Kavdlr I.Discrimination of sunflower,weed and soil by artificial neural networks.Computers and Electronics in Agriculture,2004,44:153-160.
  • 3Purcell L C.Soybean canopy coverage and light interception measurements using digital imagery.Crop Science,2000,40:834-837.
  • 4吕朝辉,陈晓光,郑元杰,吴文福,赵红霞.立体视觉技术在秧苗直立度测定中的应用[J].农业工程学报,2001,17(4):127-130. 被引量:12
  • 5Ducournau S,Feutry A,Plainchault P,Revollon P,Vigouroux B,Wagner M H.An image acquisition system for automated monitoring of the germination rate of sunflower seeds.Computers and Electronics in Agriculture,2004,44:89-202.
  • 6Granitto P M,Navone H D,Verdes P F,Ceccatto H A.Weed seeds identification by machine vision.Computers and Electronics in Agriculture,2002,33:91-103.
  • 7陈杰,李志敏,钟先信,陈文涛,刘军.烟草在线检测与异物剔除系统[J].光电工程,2003,30(5):51-54. 被引量:13
  • 8Adamsen F J,Coffelt T A,Nelson J M,Barnes E M,Rice R C.Method for using images from a color digital camera to estimate flower number.Crop Science,2000,40:704-709.
  • 9李长缨,滕光辉,赵春江,乔晓军,武聪玲.利用计算机视觉技术实现对温室植物生长的无损监测[J].农业工程学报,2003,19(3):140-143. 被引量:68
  • 10Ahmad I S,Reid J F,Paulsen M R,Sinclair J B.Color classifier for symptomatic soybean seeds using image processing.Plant Disease,1999,83(4):320-327.

二级参考文献66

  • 1郁道银,张宏,李云青,王文隽.一种新的适用于彩色图像处理的彩色空间变换方法[J].光学学报,1995,15(5):576-579. 被引量:7
  • 2科学CCD的过去、现状和未来[J].激光与光电子学进展,1995,32(10):8-10. 被引量:13
  • 3成洪,陈晓光,于海业,周云山.应用图象处理技术进行蔬菜苗特征量识别[J].农业工程学报,1995,11(4):23-26. 被引量:20
  • 4朱志刚(译).数字图像处理[M].北京:电子工业出版社,1998.127-135.
  • 5艾军 李爱民 等.北五味子不同株系间叶形指数无本质差异[J].特产研究,1999,1:43-44.
  • 6汪懋华.弘扬科学精神,推进农业工程科技创新[A].韩鲁佳.机遇,挑战,创新—2002农业工程青年科技论坛论文集[C].北京:中国农业科学技术出版社,2002.1-6.
  • 7章毓晋.图像工程[M].北京:清华大学出版社,1999.261-262.
  • 8[2]Sheffer D. Random addressable 2048× 2048 active pixel image sensor[J]. IEEE Trans. Electron Devices, 1997,44(10):1716~1720.
  • 9[3]Lrppino G. A.. Design of an 8192× 8192 pixel CCD mosaic. Proc. SPIE, 1994, 2198:810.
  • 10[4]Norbert Schuster. Geomentrical noise bandwidth:a new tool to characterize the resolving power of analogue and digital imaging devices. Proc. SPIE, 2002, 4669:336.

共引文献311

同被引文献320

引证文献32

二级引证文献327

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部