【24h】

Statistical design of experiments applied on sparkle visual detection

机译:用于闪光视觉检测的实验统计设计

获取原文
获取原文并翻译 | 示例

摘要

In this work, we have focused on analyzing how differentrnvariables affect in the process of detecting the sparkle texture effect,rnsince there are few works related to this issue. A large number ofrnsamples were available with different characteristics depending onrnthe needs of each psychophysical experiment. The studiedrnparameters were structural variables such as pigment type, size andrnconcentration, and environmental variables such as illuminationrnand viewing conditions.rnDue to the large sample number, the design of experimentsrn(DoE) was applied to optimize the psychophysical experiments andrnto evaluate which variables affect individually on the sparklerndetection, as well as if there are interactions among them. Afterrncollecting all the information from the psychophysical experiments,rnwith more than 35.000 visual assessments performed, therncorrelation with the instrumental data of the BYK-mac was verifiedrnto corroborate its reliability.rnIt has been possible to demonstrate the influence of multiscalernparameters on the sparkle appearance and to extract whichrnvariables affect to a greater extent when detecting the sparklerntexture effect. This issue was fundamental because, there was nornprevious study addressing these issues, which was of vitalrnimportance for the industry to analyze how the sample texture isrndetected depending on various conditions, whether structural,rnenvironmental, etc.rnIn addition, a reliable and applicable methodology has beenrnimplemented for other types of existing goniochromatic pigmentsrnwith novel texture effects. It has been possible to implement therndesign of experiments as a main tool to limit the visual experimentsrnto be performed, making the fewest visual evaluations possible andrnobtaining the maximum results.rnThanks to psychophysical experiments, it has been successfullyrntested the visual and instrumental correlation of sparkle detectionrnbetween the only multi-angle device on the market to date, and thernvisual assessments obtained from a large number of observers.
机译:在这项工作中,我们专注于分析变量在检测闪光纹理效果的过程中的影响,因为与此问题相关的工作很少。根据每个心理物理实验的需求,可以提供大量具有不同特征的样本。所研究的参数包括颜料类型,尺寸和浓度等结构变量,以及照明和观看条件等环境变量。由于样本数量较大,因此采用实验设计(DoE)来优化心理物理实验,并评估哪些变量会分别影响火花检测以及它们之间是否存在相互作用。在收集了来自心理物理实验的所有信息之后,进行了超过35,000次视觉评估,验证了与BYK-mac仪器数据的相关性,从而证实了其可靠性。rn有可能证明多尺度参数对火花外观的影响并提取出当检测到火花纹理效果时,哪些变量会更大程度地影响。这个问题是根本的,因为以前没有针对这些问题的研究,这对于行业分析如何根据各种条件(结构,环境等)来检测样本质地至关重要,这对行业至关重要。此外,还实施了可靠且适用的方法用于其他类型的现有角色颜料,具有新颖的纹理效果。可以将实验设计作为限制进行视觉实验的主要工具,从而使最少的视觉评估成为可能,并获得最大的结果。借助于心理物理实验,已经成功地测试了火花检测之间的视觉和仪器相关性。迄今为止,只有市场上的多角度设备,并且从大量观察者那里获得了视觉评估。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号