公开(公告)号：US20140147015A1

公开(公告)日：2014-05-29

申请号：US13684994

申请日：2012-11-26

Applicant: FRITO-LAY NORTH AMERICA, INC.

Inventor： Rick Wendell Bajema ,  Garrett Fox ,  Kevin Matthew Trick ,  David Ray Warren ,  Sheila Wright-Henry ,  Sonchai Lange ,  Wilfred Marcellien Bourg, JR.

IPC: G06K9/78

CPC classification number: G06T7/0004 ,  G06T2207/30128

Abstract: A method for scoring and controlling the quality of dynamic food products transitioning in the processing steps is performed using image analysis. An image of a plurality of moving food products on a conveyor system is captured by on-line vision equipment and image analysis is performed on the image via an algorithm that determines the percentage of pixels having varying intensities of colors and applies predetermined preferences to predict consumer dissatisfaction. The entire group of food products of one or more images is given an overall appearance score and each individual food product is also scored such that each may be ranked from least to more acceptable. The ranked food products can then be ejected in the order of worst to better rank to increase the overall quality score of the entire group.

Abstract translation: 使用图像分析进行处理步骤中动态食品转换质量的评分和控制方法。 通过在线视觉设备捕获传送系统上的多个移动的食品的图像，并且通过算法执行图像分析，该算法确定具有变化的颜色强度的像素的百分比，并应用预定的偏好来预测消费者 不满意。 给出一组或多张图像的整组食品给出总体外观评分，并且每个单独的食品也被评分，使得每个食品可以从最少到更可接受。 然后，排名最高的食品可以按最坏的顺序排出，以提高整个品牌的整体质量得分。

公开(公告)号：US09014434B2

公开(公告)日：2015-04-21

申请号：US13684994

申请日：2012-11-26

Applicant: Frito-Lay North America, Inc.

Inventor： Rick Wendell Bajema ,  Garrett Fox ,  Kevin Matthew Trick ,  David Ray Warren ,  Sheila Wright-Henry ,  Sonchai Lange ,  Wilfred Marcellien Bourg, Jr.

IPC: G06K9/00 ,  G06T7/00

CPC classification number: G06T7/0004 ,  G06T2207/30128

Abstract: A method for scoring and controlling the quality of dynamic food products transitioning in the processing steps is performed using image analysis. An image of a plurality of moving food products on a conveyor system is captured by on-line vision equipment and image analysis is performed on the image via an algorithm that determines the percentage of pixels having varying intensities of colors and applies predetermined preferences to predict consumer dissatisfaction. The entire group of food products of one or more images is given an overall appearance score and each individual food product is also scored such that each may be ranked from least to more acceptable. The ranked food products can then be ejected in the order of worst to better rank to increase the overall quality score of the entire group.

Abstract translation: 使用图像分析进行处理步骤中动态食品转换质量的评分和控制方法。 通过在线视觉设备捕获传送系统上的多个移动的食品的图像，并且通过算法执行图像分析，该算法确定具有变化的颜色强度的像素的百分比，并应用预定的偏好来预测消费者 不满意。 给出一组或多张图像的整组食品给出总体外观评分，并且每个单独的食品也被评分，使得每个食品可以从最少到更可接受。 然后，排名最高的食品可以按最坏的顺序排出，以提高整个品牌的整体质量得分。