IMAGE ANALYSIS INCLUDING TARGETED PREPROCESSING

    公开(公告)号:US20200160530A1

    公开(公告)日:2020-05-21

    申请号:US16628320

    申请日:2018-06-11

    Abstract: A system includes a K1 preprocessing module designed to generate at least one intermediate image from an input image using a parameterized internal processing chain and an analysis module to detect a feature or object in the intermediate image. A method to train the system includes feeding a plurality of learning input images to the system, comparing a result provided by the analysis module for each of the learning input images to a learning value, and feeding back a deviation obtained by the comparison to an input preprocessing module and/or adapting parameters of the internal processing chain to reduce the deviation.

    Method and device for determining a visual range in daytime fog
    7.
    发明授权
    Method and device for determining a visual range in daytime fog 有权
    用于确定白天雾中视觉范围的方法和装置

    公开(公告)号:US09506859B2

    公开(公告)日:2016-11-29

    申请号:US14777188

    申请日:2014-02-24

    Abstract: A method is provided for determining a visual range in daytime fog, the method (800) including a step of reading in and a step of ascertaining. In the step of reading in, coordinates of at least one characteristic point of a brightness curve of a camera image of the fog are read in. The brightness curve represents brightness values of image points of the camera image along a reference axis of the camera image. In the step of ascertaining, a meteorological visual range in the camera image is ascertained using the coordinates, a meteorological contrast threshold, and a processing specification, in order to estimate the visual range in fog. The processing specification images location-dependent and/or direction-dependent scattered light through the fog in the camera image.

    Abstract translation: 提供了一种用于确定白天雾中的视觉范围的方法,所述方法(800)包括读入步骤和确定步骤。 在读入的步骤中,读取雾的照相机图像的亮度曲线的至少一个特征点的坐标。亮度曲线表示摄像机图像沿摄像机图像的参考轴的图像点的亮度值 。 在确定的步骤中,使用坐标,气象对比阈值和处理规范来确定摄像机图像中的气象视觉范围,以估计雾中的视觉范围。 处理规范通过摄像机图像中的雾图像取决于位置和/或方向依赖的散射光。

    Image analysis including targeted preprocessing

    公开(公告)号:US11087470B2

    公开(公告)日:2021-08-10

    申请号:US16628320

    申请日:2018-06-11

    Abstract: A system includes a K1 preprocessing module designed to generate at least one intermediate image from an input image using a parameterized internal processing chain and an analysis module to detect a feature or object in the intermediate image. A method to train the system includes feeding a plurality of learning input images to the system, comparing a result provided by the analysis module for each of the learning input images to a learning value, and feeding back a deviation obtained by the comparison to an input preprocessing module and/or adapting parameters of the internal processing chain to reduce the deviation.

Patent Agency Ranking